Towards multiple hypothesis situation analysis

Abstract

Uncertainty makes the analysis of even simple situations difficult. It forces situation analysts to formulate and manage hypotheses during the construction of the explicit representations of the real world situations. Because of human cognitive limitations, this may quickly become overwhelming, even for the most experienced and capable analysts. In an attempt to provide better support systems, this position paper revisits the main concepts behind Multiple Hypothesis Tracking (MHT) in order to highlight how these ideas could be reused to deal with uncertainty in situation analysis. The development of a prototype reusing components of a prior MHT implementation is briefly presented. Finally, a number of challenging R&D issues and questions that have not yet been addressed are identified.